Study and Application of an Elevator Failure Monitoring System Based on the Internet of Things Technology

Scientific Programming, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 9

Published: Sept. 26, 2022

To prevent the occurrence of elevator safety accidents, in this study, an Internet things-based failure monitoring system is investigated. First, it introduces things technology, preprocesses relevant data, and extracts features operation data continuously collected by sensors. The objective paper to study application IoT elevators. Method. Relief-F algorithm used evaluate potential influencing factors. results show that, as batch size increases, accuracy rate will gradually increase, but after more than 20, decrease. When training result best. It can be seen time step prediction significantly improved. 24, highest; decline. In diagram influence learning on model, blue line indicates that 1.0, red 0.1, black 0.01. With increase iteration times, effect best when 0.1. This has high research value broad prospects make up for current lack industry.

Language: Английский

---

Positioning of Suspended Permanent Magnet Maglev Trains Using Satellite-Ground Multi-Sensor Fusion

IEEE Sensors Journal, Journal Year: 2024, Volume and Issue: 24(10), P. 16816 - 16825

Published: April 9, 2024

The reliability and continuity of maglev positioning systems are directly related to the safe operation trains. To solve problem that traditional Kalman filter is easily affected by process noise, this paper uses particle swarm optimization (PSO) algorithm improve it, proposes a model suspended permanent magnet train based on PSO-Kalman algorithm. experimental results 60 meters rail transit system technical verification line revealed compared filter, mean error (ME), relative (MRE), root square (RMSE) PSO–Kalman decreased 38.180, 23.249, 35.838%, respectively. Experiment "Redrail" Xingguo showed ME, MRE RMSE reduced 38.280, 26.994 37.027% In addition, as ability global navigation satellite (GNSS) generate signal blocking interference environment, study proposed magnetic levitation multi-sensor information fusion using GNSS, tag electronic beacon, grating axle counting information. Experiments method for single GNSS sensor, MRE, location 26.115, 19.298, 20.839%, This provides theoretical basis solving trains, an important support development transit.

Language: Английский

---

A Review on Technologies for Localisation and Navigation in Autonomous Railway Maintenance Systems

Sensors, Journal Year: 2022, Volume and Issue: 22(11), P. 4185 - 4185

Published: May 31, 2022

Smart maintenance is essential to achieving a safe and reliable railway, but traditional deployment costly heavily human-involved. Ineffective job execution or failure in preventive can lead railway service disruption unsafe operations. The of robotic autonomous systems was proposed conduct these tasks with higher accuracy reliability. In order for be capable detecting rail flaws along millions mileages they must register their location accuracy. A prerequisite an vehicle its possessing high degree terms positional awareness. This paper first reviews the importance demands networks related techniques. Furthermore, this investigates strategies, techniques, architecture, references used by different resolve network. Additionally, discusses advantages applicability on-board-based infrastructure-based sensing, respectively. Finally, analyses uncertainties which contribute vehicle’s position error influence on positioning reliability corresponding technique solutions. study therefore provides overall direction development further track-based system designs methods deal challenges faced

Language: Английский

---

Risk-Based Access Control Mechanism for Internet of Vehicles Using Artificial Intelligence

Security and Communication Networks, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 13

Published: Aug. 16, 2022

Internet of Vehicles (IoV) systems are vulnerable to a wide range attacks because the lack security measures. IoV can be infiltrated by malicious and unauthorized nodes, which cause authenticity, accessibility, privacy shared information resources compromised. Indeed, use an access control system help; as result, it is unable respond such on time. This paper introduces artificial intelligence-enabled mechanism (AI-ACM) with vehicle nodes roadside units (RSUs) overcome these issues. Here, lightweight while RSUs act comprehensive edge provide service. A generative adversarial network (GAN) used in place risk prediction (RP) due training sets, resulting sequence generation rather than accurate prediction. Afterward, blockchain-based (BIoV) approach summarized for mechanisms vehicles that discussed from aspects authentication sustain distributed processing architecture solve The simulation results show AI-ACM more previous GANs at predicting future. In addition, RP model’s accuracy improved result this technique.

Language: Английский

---

3D Model Reconstruction of CAD Virtual Reality Digital Building Based on BIM Model

Computer-Aided Design and Applications, Journal Year: 2022, Volume and Issue: unknown, P. 29 - 42

Published: Aug. 26, 2022

China has a limited share of water resources per capita and severe scarcity.It is one the world's countries with most capita.Water resource challenges in also include lack understanding conservation conservation, as well major waste, pollution, over exploitation.In order to solve problem inconvenient accurate calculation pipeline head loss complex layout, research on computer-aided design (CAD) method based Plant Design Management System (PDSM) three-dimensional model proposed.Firstly, pressure pipe taken object, straight components calculated are classified summarized according identification code component library; Secondly, formula compiled corresponding logic between parameters PDSM established; Finally, same type respectively.This approach overcomes drawbacks standard penstock method, such vast number calculations required, erroneous outputs, automation.Compared traditional technology, accuracy improved by 20%.It can calculate local synchronously, which more simpler than method.

Language: Английский

---

Research on Detection and Correction of High-Speed Train Idling and Skidding Based on Improved Fuzzy Neural Network

Automatic Control and Computer Sciences, Journal Year: 2024, Volume and Issue: 58(3), P. 274 - 288

Published: June 1, 2024

Language: Английский

---

Fault diagnosis of electrical equipment based on virtual simulation technology

Nonlinear Engineering, Journal Year: 2023, Volume and Issue: 12(1)

Published: Jan. 1, 2023

Abstract In order to efficiently and accurately diagnose train electrical faults, we propose a fault diagnosis method for equipment based on virtual simulation technology. First, Creo software was used build subway model. Then, 3DMAX make animation demonstrate the working principle action process of system. Finally, using Unity 3D software, human–computer interaction mechanism established, achieving presence realism. This system realizes functions knowledge learning, student assessment, principal display, troubleshooting trains is compared with manual diagnosis. Experimental results show that in designed system, detection time various types faults shorter than 30 s, whereas 30–52 s. It shows has advantages such as short high efficiency. addition, highest diagnostic accuracy 75.48%, which far lower Conclusion: proved can meet safety requirements industrial production.

Language: Английский

---

Three-Dimensional Numerical Simulation of Soil Deformation during Shield Tunnel Construction

Mathematical Problems in Engineering, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 11

Published: Aug. 31, 2022

In order to effectively control the ground deformation and subsidence during subway construction, three-dimensional numerical simulation of soil shield tunnel construction is proposed. Based on a project, this essay firstly divides process into several stages analyzes vertical displacement in each stage. FLAC 3D was used for finite difference simulation. By comparing results with field measured data, settlement caused by excavation studied deeply. The show that maximum value monitoring data 0.59 mm, 0.82 0.23 mm. uplift both sides 0.41 mm 0.29 simulation, 0.12 2.59 3.05 0.46 0.32 1.89 1.57 slightly larger than data. Conclusion. can well simulate state sides, width groove good agreement

Language: Английский

---

Application of Artificial Intelligence Computer Intelligent Heuristic Search Algorithm

Advances in Multimedia, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 12

Published: Sept. 24, 2022

In order to transform three-dimensional space path planning into two-dimensional plane problem and greatly reduce the search time, an intelligent heuristic algorithm based on artificial intelligence is proposed. The analyzed introduced, A chosen. spatial environment model of picking robot investigated, a established by raster method. Then, considering whole day operation time divided several periods. With help algorithm, most reasonable interval each period found, so as provide reliable reference for decision-making organization urban rail transit how arrange train rationally. experimental results show that improved ∗ can significantly improve moving make planned smoother, which confirms feasibility superiority algorithm. decision obtained through experiments. After 114 iterations optimal value 6.83353635 e -001, while average 6.83551939 e -001. 231 particle swarm optimization 6.83650785 e 6.83745935 e 789 iterations, genetic obtains 6.83921100 e 6.84410765 e Through comparative analysis, it be seen compared with other two algorithms, better than both in terms number iterations. fast, accurate, method solve accurate scheduling departure interval. It proved computer realize effectively.

Language: Английский

---

Parameter Optimization of Educational Network Ecosystem Based on BERT Deep Learning Model

Mathematical Problems in Engineering, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 9

Published: Sept. 30, 2022

The key sentimental words in the text cannot be paid attention to effectively, and language knowledge such as information resources are relied on. Therefore, it is necessary make full use of this unique achieve best performance model. In order solve problems, a method based on fusion convolutional neural network bidirectional GRU sentiment analysis capsule model analyze ideological political education public opinion put forward. model, each category combined with mechanism generate feature vectors construct capsules. Finally, categories judged according attributes tested MR, IMDB, SST-5, data set review. Experimental results show that compared MC-CNN-LSTM, readiness rate proposed improved by 5.1%, 2.8%, 1.6% four Chinese English sets, respectively. Compared LR-Bi-LSTM, NSCL, multi-Bi-LSTM models, accuracy MC-BiGRU-Capsule MR SST-5 3.2%, 2.4%, 3.4% higher than It also shows better classification effect multiclassification sets. concluded other baseline has effect.

Language: Английский

---